Method and device for processing data based on speech recognizing and scoring system

1. A method for processing data based on a speech recognizing and scoring system, comprising: acquiring at least one data frame to be processed from each of a plurality of threads to obtain a plurality of data frames; performing a frame splicing process on the plurality of data frames to obtain a spliced data frame as a target data frame; and determining a target thread from the plurality of threads, and processing the target data frame based on the target thread; wherein performing a frame splicing process on the plurality of data frames comprises: performing the frame splicing process on the plurality of data frames, such that the plurality of data frames are mutually crosswise arranged; wherein the number of frames comprised in the target data frame is equal to a value of multiplying the number of at least one data frame to be processed of each thread by the number of the plurality of threads; allocating a flag bit to the at least one data frame to be processed of each thread; and flagging the at least one data frame to be processed of each thread in the target data frame with the flag bit corresponding to the at least one data frame to be processed of the thread.

2. The method according claim 1 , after determining the target thread from the plurality of threads and processing the target data frame based on the target thread, further comprising: acquiring a processing result of the target data frame; reading, by each of threads other than the target thread, a processing result corresponding to the thread from the processing result of the target data frame based on the flag bit of the thread; and determining, by each thread, the processing result corresponding to the thread as a result of processing the at least one data frame to be processed by the thread.

3. A device for processing data based on a speech recognizing and scoring system, comprising: a processor; and a memory, configured to store one or more software modules executable by the processor; wherein the one or more software modules comprise: a first acquiring software module, configured to acquire at least one data frame to be processed from each of a plurality of threads to obtain a plurality of data frames; a frame splicing software module, configured to perform a frame splicing process on the plurality of data frames to obtain a spliced data frame as a target data frame; and a processing software module, configured to determine a target thread from the plurality of threads, and to process the target data frame based on the target thread; wherein the frame splicing software module is configured to: perform the frame splicing process on the plurality of data frames such that the plurality of data frames are mutually crosswise arranged; wherein the number of frames comprised in the target data frame is equal to a value of multiplying the number of at least one data frame to be processed of each thread by the number of the plurality of threads; an allocating software module, configured to allocate a flag bit to the at least one data frame to be processed of each thread; and a flagging software module, configured to flag the at least one data frame to be processed of each thread in the target data frame with the flag bit corresponding to the at least data frame to be processed of the thread.

4. The device according to claim 3 , wherein the one or more software modules further comprise: a second acquiring software module, configured to acquire a processing result of the target data frame; and a reading software module, configured to read, by each of threads other than the target thread, a processing result corresponding to the thread from the processing result of the target data frame based on the flag bit of the thread; and to determine the processing results corresponding to each thread as a result of processing the at least one data frame to be processed by each thread.

5. A non-transitory computer readable storage medium, having stored therein instructions that, when executed by a processor of an electronic device, cause the electronic device to perform a method for processing data based on a speech recognizing and scoring system, the method comprising: acquiring at least one data frame to be processed from each of a plurality of threads to obtain a plurality of data frames; performing a frame splicing process on the plurality of data frames to obtain a spliced data frame as a target data frame; and determining a target thread from the plurality of threads, and processing the target data frame based on the target thread; wherein performing a frame splicing process on the plurality of data frames comprises: performing the frame splicing process on the plurality of data frames, such that the plurality of data frames are mutually crosswise arranged; wherein the number of frames comprised in the target data frame is equal to a value of multiplying the number of at least one data frame to be processed of each thread by the number of the plurality of threads; allocating a flag bit to the at least one data frame to be processed of each thread; and flagging the at least one data frame to be processed of each thread in the target data frame with the flag bit corresponding to the at least one data frame to be processed of the thread.

6. The storage medium according claim 5 , wherein after determining the target thread from the plurality of threads and processing the target data frame based on the target thread, the method further comprises: acquiring a processing result of the target data frame; reading, by each of threads other than the target thread, a processing result corresponding to the thread from the processing result of the target data frame based on the flag bit of the thread; and determining, by each thread, the processing result corresponding to the thread as a result of processing the at least one data frame to be processed by the thread.